Design Strategies for Playful Technologies toSupport Light-intensity Physical Activity in the

Workplace

Misha Sra and Chris Schmandt

MIT Media Lab

Abstract. Moderate to vigorous intensity physical activity has an es-tablished preventative role in obesity, cardiovascular disease, and dia-betes. However recent evidence suggests that sitting time affects healthnegatively independent of whether adults meet prescribed physical ac-tivity guidelines. Since many of us spend long hours daily sitting in frontof a host of electronic screens, this is cause for concern. In this paper, wedescribe a set of three prototype digital games created for encouraginglight-intensity physical activity during short breaks at work. The designof these kinds of games is a complex process that must consider moti-vation strategies, interaction methodology, usability and ludic aspects.We present design guidelines for technologies that encourage physical ac-tivity in the workplace that we derived from a user evaluation using theprototypes. Although the design guidelines can be seen as general princi-ples, we conclude that they have to be considered differently for differentworkplace cultures and workspaces. Our study was conducted with userswho have some experience playing casual games on their mobile devicesand were able and willing to increase their physical activity.

Keywords: Whole-body interaction; sedentary behavior; light-intensityactivity; casual games; physical interfaces; workplace fitness;

1 Introduction

Recent evidence suggests short bouts of physical activity (i.e., standing up,stretching) are positively associated with cardiorespiratory fitness [22]. A ba-sic premise of “inactive physiology” is that sitting too much is not the same aslack of exercise and has its own metabolic consequences [11]. With that in mind,we try to answer the question of what an individual who sleeps eight hours aday and exercises for 30 minutes, can do the rest of the 15.5 hours of their daythey are not exercising in order to improve their health and wellbeing. Studiesshow what people do in their non exercise time may be equally beneficial forhealth as 30 minutes of recommended focused exercise activity [11]. Substantialresearch has been devoted to promoting physical activity using mobile devicesthrough activity trackers and sensors that calculate the number of steps walked,distance run, calories expended, or heart rate achieved. These systems employadditional approaches like virtual rewards, game elements, social incentives, and

arX

iv:1

512.

0292

1v1

[cs

.HC

] 9

Dec

201

5

2 Misha Sra and Chris Schmandt

goal setting to motivate users to be more active with varying degrees of suc-cess. However, there is limited prior research on promoting physical activity [18]during work breaks and we believe this is a missed opportunity for beneficiallyusing time to ameliorate the adverse health effects of sedentary behavior. Thegrowing prevalence of stand-up and treadmill desks shows a desire for creativelyaddressing the perils of sitting. Our target audience is graduate students andstaff in a research university environment. It is generally believed that goodhabits learned during childhood stay with us for life. Following from that belief,we hope to create new exercise habits in our peer group population so that whenwe graduate and juggle work and family responsibilities, we don’t neglect ourwellbeing and good habits learned during graduate school stay with us.

We believe providing a digital interface to familiar physical actions with gameelements like goals, challenges, levels, rewards, and high scores could add mo-tivation for frequent use necessary for impacting sedentary behavior. Thus, wedesigned three physically interactive digital games: See-Saw, a balancing gamefor Google Glass; Jump Beat, a music beat matching game for Google Glass;and Learning To Fly, a Microsoft Kinect game where two users fly a virtualbird through obstacles by flapping their arms in sync. In addition to whole-body movement, we capture heart rate patterns by attaching a pulse sensor toGlass to make a connection between the user’s physiological state and their con-text of interaction. Each game is designed to explore different user motivationsfrom free-form self directed play in See-Saw to guided play in Jump Beat tocooperative play in Learning to Fly to help create a comprehensive set of designguidelines. We evaluated prototypes of the three games with a group of users whowanted to increase their levels of physical activity. The results of the user studyare being used to inform the design of new applications to enable a longitudinalstudy for examining impact on sedentary behavior.

In this paper, we focus our discussion on the four key design elements fortechnologies that encourage physical activity which we derived from our anal-ysis of the qualitative data collected from the user study: 1) Provide personalawareness of current activity state, 2) Use simple game mechanics for mappingphysical actions to virtual elements, 3) Support social interactions, and 4) Con-sider the practical constraints of users’ workplace culture, space, interests, andability. The remainder of this paper is organized as follows: First, we presentrelated work. Next, we describe the design of the three games. We follow the de-sign with the user study methodology and discuss the results. Finally, we presentthe design guidelines and conclude with a summary of our findings.

2 Related Work

Our work is informed by prior work in three distinct areas: depth and sensorbased approaches for physical interaction; movement-based play; and short du-ration breaks at work. Research has shown exergames, which combine exerciseand digital games motivate people to be more physically active [19]. Our goal isto ameliorate the adverse effects of extended periods of sitting, an unavoidable

Title Suppressed Due to Excessive Length 3

consequence of modern work and lifestyles, using playful mechanisms. Increas-ing bodily movement by taking breaks from sitting is advocated as a means forincreasing metabolism [12] and reducing the risk of cardiovascular disease [15].Recent evidence suggests health consequences of “too much sitting” are distinctfrom those of “too little exercise” [11]. Substantial amount of research has beendevoted to promoting physical activity in general through mobile devices withembedded sensors [14,9] or through dedicated devices like the Nike Fuel Band1

or the FitBit2 to calculate number of steps walked, distance run, or calories ex-pended. However, there is limited work that promotes physical activity duringprolonged sitting at work and we hope to fill this gap.

Physical Interaction In Brave NUI World, Wigdor and Wixon describe anatural user interface as a combination of UI and experiences created by lever-aging the potential of new technologies to better mirror human capabilities [24].Physical interactions for manipulating digital information using motion-basedcontrollers are an example of natural interaction with technology. The recentKinect Sports Rivals3 is an example game that scans and creates a stylized dig-ital version of the user. It uses body movements for interaction and reflects theplayer’s expressions on their avatar’s face. More recently, free-form hand gesturalinteraction with head-mounted displays (HMDs) is made possible with the Mimesensor [5] while the Leap Motion Controller4 allows hand and finger gestures asinput for laptop devices. A variety of interactive systems are feasible with gestu-ral interaction like creating animations with KinEtre [4] recording and learningphysical movement sequences with YouMove [1], and supporting immersive videosee-through augmented reality (AR) with the Oculus Rift5 in AR-Rift6. We takeinspiration from natural user interfaces in the design of gestural and whole bodymovement-based interaction for our games.

Movement-based Play Beyond encouraging social connection, play consti-tutes an emotionally significant context that can transform tedious physical ac-tivity into an enjoyable experience. Movement-based play puts the body in thecenter of the interactive experience where gross-motor bodily input influencesoutcomes in the digital world [20]. Through play, exploration or other similaractivities people experience positive emotions with benefits to physical, intellec-tual, and social well-being. [2,21]. Though a relatively recent phenomenon, thereexists evidence to suggest the potential of exergames to encourage physical ac-tivity [3,25]. Commercially available games like Dance Dance Revolution havebeen found to increase aerobic fitness [10] and similarly WiiFit has been foundto be more enjoyable than traditional exercise [17]. Exergames in research have

1 Nike. http://www.nike.com/us/en us/c/nikeplus-fuelband2 Fitbit. http://www.fitbit.com3 Kinect. http://www.xbox.com/en-US/xbox-one/games/kinect-sports-rivals4 Leap Motion. https://www.leapmotion.com/5 Oculus VR. http://www.oculusvr.com6 AR-Rift. http://willsteptoe.com/post/66968953089/ar-rift-part-1

4 Misha Sra and Chris Schmandt

(a) See-Saw on Glass. (b) Jump Beat midi sequence.

Fig. 1: (a): Glass screen (inset) showing ball 3 of 3 about to fall off the beam asthe user tries to balance themselves playing See-Saw. (b): Partial midi sequenceof a song used in Jump Beat showing drums in orange and electric bass in yellowin MidiEditor on Windows 7.

integrated standard exercise equipment like bikes with play [25] and used phys-iological sensing like player’s heart-rate for controlling the game [7]. The designof our prototype games uses a combination whole-body actions and physiologicalsignals as input for the digital systems.

Short Duration Breaks Research studies have looked at break-reminder soft-ware as a means of improving voluntary break-taking with positive effects onproductivity and well-being [13]. Lin et al. [16] have used personal awarenessand social pressure to increase daily walking while Morris et al. have used phys-ical interactions to motivate good ergonomic habits [18] in SuperBreak. In astudy of SitCoach [6] it was shown that persuasive messages significantly re-duced computer activity, however, no significant changes in physical activitywere observed. Our work focuses on the design of short duration digital gamesto encourage physical activity during work breaks to reduce overall sitting time.The length of short breaks mostly depends on the individual and the workplaceculture though some recommendations suggest taking a three - five minutesbreak 25 minutes e.g. the Pomodoro technique 7.

3 Implementation

Our prototypes are composed of a piece of commercial hardware–Google Glassor Microsoft Kinect–and our custom software that runs on each. In this section,we describe the three applications that were used in the user study.

See-Saw on Google Glass In See-Saw, the user is tasked with balancing avirtual ball on a beam (see Figure 1a). There is no time, equipment, and overhead

7 http://pomodorotechnique.com

Title Suppressed Due to Excessive Length 5

Fig. 2: Three Glass screenshots showing sequence of progression as user missesone scrolling note (middle screenshot) and matches the next one (right screen-shot) in Jump Beat. User score and BPM visible in each screenshot.

required to setup the system which makes it convenient for people to interrupttheir day with multiple short bursts of activity by simply putting on Glass andusing voice commands to start the See-Saw application. See-Saw has 2D (twodimensional) visuals and is written in Java using the Glass Development Kit(GDK) and the Android SDK. The simple design and free-form interaction allowsusers to challenge themselves by incorporating rules on the fly like standing onone foot, walking, or jumping while trying to balance the ball. The embeddedaccelerometer and gyroscope in Glass provide data for calculating the orientationof the user’s head. The returned angle value is used for animating the blue beamon screen. We used the Jbox2D8 physics engine for bringing the experience closerto that of balancing a real ball. The physical input from the user tilting theirhead left or right keeps the beam level and prevents the ball from falling off thebeam. The sensitivity of the simulated physics can be adjusted to provide aneasier or more challenging experience.

Jump Beat on Google Glass The game requires users to jump in time tomatch music notes that scroll on-screen in order to score points (see Figure 2).The visuals are 2D and the game takes inspiration from an older console gameGuitar Hero9. The frequency of notes is affected by the user’s heart rate which ismeasured by a pulse sensor. When the heart rate goes over a pre-set threshold,the frequency of notes slows down allowing the user to catch their breath and ifit goes below the threshold, the frequency of notes goes up requiring the user tojump often to keep up with the game. The notes are matched to drum beats inthe song and for our prototype we created one playable song. In order to matchthe notes on screen to the beats in the song we converted the song MP3 to midi(see Figure 1b) and manually analyzed the result for beat frequency. Based onresulting frequency we generated a map to use in our software for synchronizingthe animation to the beats.

Data from a pulse sensor is employed for providing input into the system inaddition to the “jumping” action by the user (see Figure 3a). The sensor inputis dependent on the “jumping” action which is used in two ways for affectingthe system: directly as a physical interaction mechanic such that jumping is

8 A Java port of the Box2D physics engine. http://www.jbox2d.org/9 Guitar Hero. http://www.guitarhero.com/

6 Misha Sra and Chris Schmandt

(a) Jump Beat on Glass.(b) Learning to Fly on Kinect.

Fig. 3: (a): user jumping to match scrolling notes visible on Glass screen (seeFigure 2). The pulse sensor is held between the right index finger and thumband connect to Glass through the IOIO board. Figure (a) top right: pulse sensorconnected to IOIO board. Figure (a) bottom right: Glass connected to pulsesensor through the IOIO board. (b): one person playing Learning To Fly byflapping both arms. Figure (b) top right: screenshot of Learning to Fly showingplayer high score and bird flying through obstacles. Figure (b) bottom right: twousers interacting with Learning To Fly by flapping their arms together.

equivalent to a button press of a game controller or a key press on the keyboard;indirectly as heart rate data that affects the frequency of music notes scrollingon the screen. The heart rate calculated from the pulse sensor depends on howmuch the user is jumping and how the jumping is affecting them physiologically.The pulse sensor can be held between the thumb and index finger or clipped toan ear lobe to provide hands-free sensor data. We found the sensor to be moresensitive when clipped to the ear than when held in the hand for some users.Holding the pulse sensor too hard can squeeze all the blood out of the fingertipleading to poor or no signal while holding it too lightly can lead to noise frommovement and ambient light. The pulse sensor10 is connected to Glass throughthe IOIO board11 over USB (see Figure 3a).

Learning To Fly on Microsoft Kinect The game was designed to inviteparticipation from officemates or friends for a collaborative and social work breakbut it can also be played alone (see Figure 3b). The game is a 2D clone ofa currently popular mobile game “Flappy Bird” but instead of touch basedinteraction it uses movement-based input to keep a virtual bird in flight. Theonscreen bird moves from the left side of the screen to the right and encounters

10 We used a commercially available pulse sensor fromhttps://www.sparkfun.com/products/11574

11 IOIO. https://github.com/ytai/ioio/wiki

Title Suppressed Due to Excessive Length 7

series of walls with openings at varying heights that it needs to go through. If itflies into a wall the game starts over. In the two player mode, both people needto flap one arm each and stay synchronized for keeping the bird in the air. Thegame is written in C# using the Microsoft Kinect Toolkit and the Microsoft XNAFramework with the Farseer Physics Engine. The Kinect sensor is used to getskeleton and joint data for users standing in front of the sensor. A flap is definedas a cycle of movement where both arms go up, down, and back up togetherfor one user. Two players use one opposite arm each to flap synchronously (seeFigure 3b). The position and movement over time of the hand joint in each arm,relative to the respective elbow and shoulder joints is used to calculate the flap.

Both hands for each user (Kinect provides tracking for a maximum of twoskeletons while detecting up to 6) were independently tracked to allow the systemto be flexible and usable by one person or two people. The design allows fora second player to fluidly enter or exit an ongoing one player game withoutrequiring a restart of the game. The XNA platform provided the physics engineused to simulate gravity in the game as well as create the canvas for displayinganimations. The interaction involved alternately flapping and letting go in orderto guide the bird through gaps in walls that occur at varying heights.

4 Evaluation

We conducted an experimental user study with eleven volunteer participants(ages 25–44, 5F, mean age: 31.09, standard deviation: 6.52) to gain useful feed-back from end users. Three users played only See-Saw while eight played JumpBeat and Learning to Fly. To learn about social aspects and acceptance of phys-ical activity in open or shared workspaces we brought the devices to the users’offices instead of conducting a lab study. Participants were asked beforehandif they would comfortably be able to do the physical actions required in eachgame since our current system is targeted at users who are willing and able todo the physical actions. Before each session, users were introduced to the goaland the physical action required to interact with each game. Participants filledout a post-study questionnaire, with ten of the fourteen items consisting of afive-point Likert scale with two bipolar anchors to mark the opposing ends ofthe scale (never - frequently, frustrating - fun). The rest of the questions askedwhether they enjoyed the game, how it compared to their existing break activi-ties, whether the physical interaction was engaging, and how it felt playing in ashared workspace. In addition to estimating the overall value of physically inter-active breaks, we assessed the subjective perception of each of the three games,to inform the design of future physical interaction based systems. Each test ses-sion comprised of participants playing the game for two to five minutes followedby 25 minutes12 of work, during which they were encouraged to continue as theywould on a normal work day. At the 25 minute mark they were asked to playagain. We repeated this over time to include three breaks and four play sessions.Since for several participants, this was the first time they had used Glass, we

12 Based on the Pomodoro technique http://pomodorotechnique.com/

8 Misha Sra and Chris Schmandt

tried to counter the novelty effect by asking participants to play multiple times.Still some of the success of See-Saw and Jump Beat may partially be attributedto the novelty of using new technology.

4.1 Results

Regarding their experience, 100% found playing a game fun and engaging eventhough only two out of eleven admitted to playing casual games on their mobiledevices frequently. Ten participants found the amount of physical activity persession “just right” while one person did not think it was even noticeable. Theactivities were rated enjoyable with a median response of 4 and variance 0.73 ona scale of 1 to 5 where 5 meant extremely enjoyable. The physical interaction wasmore enjoyable than keyboard based interaction as confirmed by the Wilcoxontest13 with p ¡ 0.05, indicating a trend and not mere coincidence in the data.87.5% of the users enjoyed jumping to music, 100% enjoyed the social interaction,and 66.66% enjoyed improvising in See-Saw. The effect of the heart rate sensoron the game was not noticeable to most users but people enjoyed seeing theirBPM (beats per minute) values. Two participants did not notice anything onthe Glass display other than the scrolling notes and that may be due to thesmall size of the screen and the focus required to coordinate jumping with thescrolling notes. 87.5% participants specifically mentioned enjoying the activityon Glass because of the hands-free experience. Almost all participants said theywould like to integrate physical games and other activity into their break time.Six participants shared offices with one other person while one participant hadtwo officemates. Others worked in an open space.

4.2 Key Design Guidelines

Several themes emerged from our analysis of the qualitative data, which wepresent as four key design guidelines of technologies that encourage physicalactivity in the workplace:

1. Provide personal awareness of current activity state,2. Use simple game mechanics for mapping physical actions to virtual elements,3. Support social interactions,4. Consider the practical constraints of users’ workplace culture, space, inter-

ests, and ability. lifestyles.

Based on user evaluation we formulated a set of design guidelines for thedevelopment of systems for introducing light-intensity physical activity into oursedentary work lives. We believe these guidelines are applicable to the designof movement-based casual game systems in general and expand upon existingcasual game design guidelines [8]. Understanding game play, social interaction,and spectatorship also provide valuable input to the design of our physicallyinteractive systems for semi-private workspaces.

13 Wilcoxon Test. http://en.wikipedia.org/wiki/Wilcoxon_signed-rank_test

Title Suppressed Due to Excessive Length 9

Use simple game mechanics for mapping physical actions to virtualelements Most video games are designed for dedicated play. Games that canbe played during short breaks at work need to be quick and easy to learn withfast reachable goals similar to casual games14 on mobile devices. This is alsoimportant for multiplayer games where the game may be concurrent with otheractivities like social interaction. In addition, identifying the right kind of chal-lenges for addressing different skill levels and providing an evolving experienceto keep the game interesting are valuable design considerations. All participantssaid the games were “quick fun” and a few remarked “jumping and flapping feltnatural.” One user found the lack of explicit game mechanics in See-Saw con-fusing and wanted “instructions to show on screen.” Most of the users said theyliked playing a physical version of a digital game they were familiar with say-ing they “already know how to play” and “I can get good at it fast.” Mappingphysical input with corresponding digital functions correctly is important forcreating intuitive user experiences. For example, asking the user to move theirhead side to side in order to make a virtual bird fly would not be as natural asasking them to flap their arms. Another valid strategy would be to select familiarphysical movements and connecting the action with things that are universallycompelling like music in Jump Beat. Additionally, similar to traditional games,if designing a multiplayer game, it should allow people to join or leave the gameat any point without adversely affecting the game experience.

Provide personal awareness of current activity state In movement-basedinteraction, the body is the major focus of attention [19]. The design of the vi-sual interface should be simple and not distract from this focus on the body andprovide enough information such that the user is able to make a connection be-tween their physical action and it’s virtual counterpart. The user should be ableto view the visual representation on screen, consider how to affect the elementsin the represented system and then perform the physical action necessary forthe manipulation. The system, in turn, should translate the physical movementand/or sensor data into an input request, update itself by changing game stateand present the updated state to the user. The combination of feedback andclear interaction with the system make for a smooth and engaging experience. Auser playing Jump Beat remarked on the lack of visual feedback on their jumpswhile another user wanted more “visual flourishes” added to the interface tomake it even more game like. One user found See-Saw boring and was uncertainabout how to play if it required making their own rules. Two other users foundthe lack of rules interesting and wanted to continue challenging themselves butwanted their running score displayed on screen in addition to the score for eachlevel. One See-Saw user wanted “more feedback” and “more points” and saidthe bright colors helped them “focus on the screen.”

Support social interactions An important class of influence that had im-pact on motivation was the opportunity for social interaction during break time.

14 http://en.wikipedia.org/wiki/Casual_game

10 Misha Sra and Chris Schmandt

Physically active play with a social component is engaging, motivating, and de-sirable. It also works better for people in shared workspaces than single playerphysical interaction for allowing people to take shared breaks which are less dis-ruptive since everyone in the space is engaged in a shared activity. The physicaland social interactions were noted as factors that made the games more enter-taining. A number of participants commented that they liked the idea of “socialphysical activity” which made “taking a break more exciting.” Six Learning ToFly participants said, “the activity encouraged human contact” and playing withsomeone made it more enjoyable. Findings from the questionnaire give the aver-age level of excitement rated by participants as 4.23 (variance of .64), signifyinga high level of excitement while doing a physical action together with others. Forgames with a visible display screen, spectating is an important element of theexperience allowing for onlookers to transition into players and back seamlessly.We believe this ease of movement between spectatorship and playing and theoverall social aspect of playing together can be a strong motivator for people toengage in physical activity more frequently.

Consider the practical constraints of users’ workplace culture, space,interests, and ability. lifestyles 81.8% of the users felt comfortable usingGlass and the heart rate sensor and some remarked that it made them “lookcool” and liked having a ”private screen.” Furthermore, 90.9% of the playersfound the hands-free interaction enjoyable especially in Jump Beat and Learn-ing to Fly. One user said they felt “much more in the game” and it is “more funwhen it’s on your head” about playing on Glass. Four users felt self-consciouswhile jumping alone in Jump Beat but all eight felt comfortable while flappingtheir arms together with another person in Learning to Fly, regardless of thetype of work space. Though mobile phone use, especially talking and ringing inpublic spaces has blurred the boundaries between public and private behaviors[23], the same is not true of digital activity in the work place. Therefore, an un-derstanding between people sharing office space regarding acceptable behaviorswould be an important design consideration for a digital system that requiresphysical actions as input. Another aspect of physical interaction in a sharedworkspace is social embarrassment which may be overcome by creating systemsthat invite participation from others instead of putting one individual on dis-play or creating systems for shared break rooms. Those participants who playedJump Beat in a shared office were conscious of their jumping even though theirofficemates ignored their jumping. Those who attempted Jump Beat in a privateoffice or Learning to Fly in an open shared workspace felt comfortable doing thephysical actions and did not think they were bothering anyone by their activity.Different work culture may allow for different acceptable behaviors for e.g. in anacademic research lab it may be acceptable to jump or do pushups in an openwork space which may not be the case in a formal corporate environment unlessthe individual has a private office. Similarly, people dress casually for formallydepending on their work place and system design needs to take that into con-

Title Suppressed Due to Excessive Length 11

sideration along with target audience specifics like physical activity willingnessand ability while designing interactions.

5 Conclusion

Based on the response to our set of physically interactive games we are ex-cited about their potential to help combat the health hazards of sitting for longstretches. Though physical interactions and play were reported as exciting andfun, they are potentially disruptive in shared offices which could severely limittheir regular use. We believe there is room for physical games in the work envi-ronment, especially social play in common areas or solo play on mobile deviceslike Glass for outdoors as well as private office spaces. What pleasantly sur-prised us was the extent to which some participants wanted to continue playingbeyond the duration of a “short” break or start playing before it was time fora break which supports our belief that intense blasts of engaging play can mo-tivate people into being physically active. Though qualitative research can haveless statistical power than quantitative research when it comes to discoveringand verifying trends, our small user study has given us broad guidelines for thedesign of other such games. Using that information, we can design newer proto-types that can be deployed to a larger set of people for gathering more data forassessing impact on sedentary lifestyles.

References

1. Anderson, F., Grossman, T., Matejka, J., and Fitzmaurice, G. Youmove: En-hancing movement training with an augmented reality mirror. In UIST (2013),311–320.

2. Aspinall, P., Mavros, P., Coyne, R., and Roe, J. The urban brain: analysing out-door physical activity with mobile eeg. British journal of sports medicine (2013),bjsports–2012.

3. Bianchi-Berthouze, N., Kim, W., and Patel, D. Body movement engage you morein digital game play? and why? In Affective Comp. and Intelligent Interaction(2007), 102–113.

4. Chen, J., Izadi, S., and Fitzgibbon, A. KinEtre: Animating the world with thehuman body. In UIST, ACM (2012), 435–444.

5. Colaco, A., Kirmani, A., Yang, H. S., Gong, N.-W., Schmandt, C., and Goyal,V. K. Mime: Compact, low power 3d gesture sensing for interaction with headmounted displays. In UIST (2013), 227–236.

6. Dantzig, S., Geleijnse, G., and Halteren, A. T. Toward a persuasive mobile ap-plication to reduce sedentary behavior. Personal Ubiquitous Comput. 17, 6 (Aug.2013), 1237–1246.

7. de Oliveira, R., and Oliver, N. Triplebeat: Enhancing exercise performance withpersuasion. In Proc. MobileHCI, ACM (2008), 255–264.

8. Deterding, S., Dixon, D., Khaled, R., and Nacke, L. From game design elementsto gamefulness: defining gamification. In Proceedings of the 15th InternationalAcademic MindTrek Conference: Envisioning Future Media Environments, ACM(2011), 9–15.

12 Misha Sra and Chris Schmandt

9. Fujiki, Y. iphone as a physical activity measurement platform. In CHI EA (2010),4315–4320.

10. Gao, Z., Zhang, T., and Stodden, D. Children’s physical activity levels and psy-chological correlates in interactive dance versus aerobic dance. In J Sport HealthSci. (2013), 146–151.

11. Hamilton, M. T., Healy, G. N., Dunstan, D. W., Zderic, T. W., and N., O. Toolittle exercise and too much sitting: Inactivity physiology and the need for newrecommendations on sedentary behavior. In Curr Cardiovasc Risk Rep. (2008),292–298.

12. Healy, G. N., Dunstan, D. W., Salmon, J., Cerin, E., Shaw, J. E., Zimmet, P. Z.,and Owen, N. Breaks in sedentary time beneficial associations with metabolic risk.Diabetes care 31, 4 (2008), 661–666.

13. Henning, R., Jacques, P., Kissel, G., Sullivan, A., and Alteras-Webb, S. Frequentshort rest breaks from computer work: effects on productivity and well-being attwo field sites. In Ergonomics (1997), 78–91.

14. Klasjna, P., Consolvo, S., McDonald, D., Landay, J., and Pratt, W. Using mobile& personal sensing technologies to support health behavior change in everyday life:lessons learned. In Amer. Med. Info. Assoc. (2009).

15. Lee, I. M., Shiroma, E. J., Lobelo, F., Puska, P., Blair, S. N., and Katzmarzyk,P. T. Effect of physical inactivity on major non-communicable diseases worldwide:an analysis of burden of disease and life expectancy. The Lancet 380, 9838 (2012),219–229.

16. Lin, J. J., Mamykina, L., Lindtner, S., Delajoux, G., and Strub, H. B. Fish’n’steps:Encouraging physical activity with an interactive computer game. In Proc. UBI-COMP (2006), 261–278.

17. Lyons, E., D.F., T., Ward, D., Bowling, J., Ribisl, K., and Kalyararaman, S. Energyexpenditure and enjoyment during video game play: differences by game type. InMed Sci Sports Exerc., vol. 43 (2011), 1987–93.

18. Morris, D., Brush, A. B., and Meyers, B. R. Superbreak: Using interactivity toenhance ergonomic typing breaks. In Proc. CHI (2008), 1817–1826.

19. Mueller, F., and Agamanolis, S. Sports over a distance. In CIE (2005), 3–4.20. Mueller, F., Edge, D., Vetere, F., Gibbs, M., Agamanolis, S., Bongers, B., and

Sheridan, J. Designing sports: A framework for exertion games. In Proc. CHI(2011), 2651–2660.

21. Mueller, F., and Isbister, K. Movement-based game guidelines. In Proceedings ofthe 32nd annual ACM conference on Human factors in computing systems, ACM(2014), 2191–2200.

22. Ross, R., and McGuire, A. Incidental physical activity is positively associated withcardiorespiratory fitness. Medicine & Science in Sports & Exercise 44, 3 (2012),2189–2194.

23. Wei, R., and Leung, L. Blurring public and private behaviors in public space:policy challenges in the use and improper use of the cell phone. Telematics andInformatics 16, 1-2 (1999), 11–26.

24. Wigdor, D., and Wixon, D. Brave NUI world [electronic resource] : designingnatural user interfaces for touch and gesture / Daniel Wigdor, Dennis Wixon.Burlington, Mass. : Morgan Kaufmann, c2011 (Norwood, Mass. : Books24x7.com),2011.

25. Yim, J., and Graham, T. C. N. Using games to increase exercise motivation. InProc. Future Play, ACM (2007), 166–173.